switch the documentation to run with numpy>=2 (pydata#9177)

* `NaN` → `nan`

* new numpy scalar repr

* unpin `numpy` in the docs [skip-ci]

* try extracting the scalars to pass to `linspace`

* use `numpy` instead of `numpy.array_api` [skip-ci]

and mention that we're falling back to `numpy.array_api` in `numpy<2.0`

* Update ci/requirements/doc.yml

* Remove 2D cross product doctests

* One more fix

---------

Co-authored-by: Jessica Scheick <JessicaS11@users.noreply.github.com>
Co-authored-by: Deepak Cherian <dcherian@users.noreply.github.com>
Co-authored-by: Deepak Cherian <deepak@cherian.net>

Author: 3 people

ci/requirements/doc.yml

@@ -21,7 +21,7 @@ dependencies:
   - nbsphinx
   - netcdf4>=1.5
   - numba
-  - numpy>=1.21,<2
+  - numpy>=2
   - packaging>=21.3
   - pandas>=1.4,!=2.1.0
   - pooch

doc/getting-started-guide/faq.rst

@@ -352,9 +352,9 @@ Some packages may have additional functionality beyond what is shown here. You c
 How does xarray handle missing values?
 --------------------------------------
 
-**xarray can handle missing values using ``np.NaN``**
+**xarray can handle missing values using ``np.nan``**
 
-- ``np.NaN`` is  used to represent missing values in labeled arrays and datasets. It is a commonly used standard for representing missing or undefined numerical data in scientific computing. ``np.NaN`` is a constant value in NumPy that represents "Not a Number" or missing values.
+- ``np.nan`` is  used to represent missing values in labeled arrays and datasets. It is a commonly used standard for representing missing or undefined numerical data in scientific computing. ``np.nan`` is a constant value in NumPy that represents "Not a Number" or missing values.
 
 - Most of xarray's computation methods are designed to automatically handle missing values appropriately.
 

doc/user-guide/computation.rst

@@ -426,7 +426,7 @@ However, the functions also take missing values in the data into account:
 
 .. ipython:: python
 
-    data = xr.DataArray([np.NaN, 2, 4])
+    data = xr.DataArray([np.nan, 2, 4])
     weights = xr.DataArray([8, 1, 1])
 
     data.weighted(weights).mean()
@@ -444,7 +444,7 @@ If the weights add up to to 0, ``sum`` returns 0:
 
     data.weighted(weights).sum()
 
-and ``mean``, ``std`` and ``var`` return ``NaN``:
+and ``mean``, ``std`` and ``var`` return ``nan``:
 
 .. ipython:: python
 

doc/user-guide/interpolation.rst

@@ -292,8 +292,8 @@ Let's see how :py:meth:`~xarray.DataArray.interp` works on real data.
     axes[0].set_title("Raw data")
 
     # Interpolated data
-    new_lon = np.linspace(ds.lon[0], ds.lon[-1], ds.sizes["lon"] * 4)
-    new_lat = np.linspace(ds.lat[0], ds.lat[-1], ds.sizes["lat"] * 4)
+    new_lon = np.linspace(ds.lon[0].item(), ds.lon[-1].item(), ds.sizes["lon"] * 4)
+    new_lat = np.linspace(ds.lat[0].item(), ds.lat[-1].item(), ds.sizes["lat"] * 4)
     dsi = ds.interp(lat=new_lat, lon=new_lon)
     dsi.air.plot(ax=axes[1])
     @savefig interpolation_sample3.png width=8in

doc/user-guide/testing.rst

@@ -239,9 +239,10 @@ If the array type you want to generate has an array API-compliant top-level name
 you can use this neat trick:
 
 .. ipython:: python
-    :okwarning:
 
-    from numpy import array_api as xp  # available in numpy 1.26.0
+    import numpy as xp  # compatible in numpy 2.0
+
+    # use `import numpy.array_api as xp` in numpy>=1.23,<2.0
 
     from hypothesis.extra.array_api import make_strategies_namespace
 

xarray/core/computation.py

@@ -1064,7 +1064,7 @@ def apply_ufunc(
     supported:
 
     >>> magnitude(3, 4)
-    5.0
+    np.float64(5.0)
     >>> magnitude(3, np.array([0, 4]))
     array([3., 5.])
     >>> magnitude(array, 0)
@@ -1587,15 +1587,6 @@ def cross(
     array([-3,  6, -3])
     Dimensions without coordinates: dim_0
 
-    Vector cross-product with 2 dimensions, returns in the perpendicular
-    direction:
-
-    >>> a = xr.DataArray([1, 2])
-    >>> b = xr.DataArray([4, 5])
-    >>> xr.cross(a, b, dim="dim_0")
-    <xarray.DataArray ()> Size: 8B
-    array(-3)
-
     Vector cross-product with 3 dimensions but zeros at the last axis
     yields the same results as with 2 dimensions:
 
@@ -1606,42 +1597,6 @@ def cross(
     array([ 0,  0, -3])
     Dimensions without coordinates: dim_0
 
-    One vector with dimension 2:
-
-    >>> a = xr.DataArray(
-    ...     [1, 2],
-    ...     dims=["cartesian"],
-    ...     coords=dict(cartesian=(["cartesian"], ["x", "y"])),
-    ... )
-    >>> b = xr.DataArray(
-    ...     [4, 5, 6],
-    ...     dims=["cartesian"],
-    ...     coords=dict(cartesian=(["cartesian"], ["x", "y", "z"])),
-    ... )
-    >>> xr.cross(a, b, dim="cartesian")
-    <xarray.DataArray (cartesian: 3)> Size: 24B
-    array([12, -6, -3])
-    Coordinates:
-      * cartesian  (cartesian) <U1 12B 'x' 'y' 'z'
-
-    One vector with dimension 2 but coords in other positions:
-
-    >>> a = xr.DataArray(
-    ...     [1, 2],
-    ...     dims=["cartesian"],
-    ...     coords=dict(cartesian=(["cartesian"], ["x", "z"])),
-    ... )
-    >>> b = xr.DataArray(
-    ...     [4, 5, 6],
-    ...     dims=["cartesian"],
-    ...     coords=dict(cartesian=(["cartesian"], ["x", "y", "z"])),
-    ... )
-    >>> xr.cross(a, b, dim="cartesian")
-    <xarray.DataArray (cartesian: 3)> Size: 24B
-    array([-10,   2,   5])
-    Coordinates:
-      * cartesian  (cartesian) <U1 12B 'x' 'y' 'z'
-
     Multiple vector cross-products. Note that the direction of the
     cross product vector is defined by the right-hand rule:
 

xarray/plot/facetgrid.py

@@ -774,7 +774,7 @@ def _get_largest_lims(self) -> dict[str, tuple[float, float]]:
         >>> ds = xr.tutorial.scatter_example_dataset(seed=42)
         >>> fg = ds.plot.scatter(x="A", y="B", hue="y", row="x", col="w")
         >>> round(fg._get_largest_lims()["x"][0], 3)
-        -0.334
+        np.float64(-0.334)
         """
         lims_largest: dict[str, tuple[float, float]] = dict(
             x=(np.inf, -np.inf), y=(np.inf, -np.inf), z=(np.inf, -np.inf)
@@ -817,7 +817,7 @@ def _set_lims(
         >>> fg = ds.plot.scatter(x="A", y="B", hue="y", row="x", col="w")
         >>> fg._set_lims(x=(-0.3, 0.3), y=(0, 2), z=(0, 4))
         >>> fg.axs[0, 0].get_xlim(), fg.axs[0, 0].get_ylim()
-        ((-0.3, 0.3), (0.0, 2.0))
+        ((np.float64(-0.3), np.float64(0.3)), (np.float64(0.0), np.float64(2.0)))
         """
         lims_largest = self._get_largest_lims()
 

xarray/plot/utils.py

@@ -811,11 +811,11 @@ def _update_axes(
 def _is_monotonic(coord, axis=0):
     """
     >>> _is_monotonic(np.array([0, 1, 2]))
-    True
+    np.True_
     >>> _is_monotonic(np.array([2, 1, 0]))
-    True
+    np.True_
     >>> _is_monotonic(np.array([0, 2, 1]))
-    False
+    np.False_
     """
     if coord.shape[axis] < 3:
         return True